Measurement problems are important to the HVAC industry. Measurement disputes are often at the heart of conflicts over HVAC performance issues such as uncomfortable buildings, inefficient energy performance, and inability to maintain specified parameters such as adequate positive pressure in hospital operating rooms. These conflicts frequently result in anger, confusion, disputes, cancelled contracts, lawsuits, mediation, and unhappy building owners, tenants, and workers. Contributing to these conflicts is that measurements of HVAC-related building parameters such as air and water temperature, humidity, pressure, velocity, and flow are perceived to be inaccurate and unreliable, so dissatisfied parties often challenge their validity.
Fans create pressure differences, which force air to flow through the duct system of a building. Fans in the air handling unit (AHU) generate the energy necessary to overcome the duct system's resistance to airflow. Resistance is offered by filters and heat exchange coils. Straight ducts have resistance in proportion to their length. Bends and size changes increase resistance to flow. Diffusers (grilles, outlets) offer the final resistance before the air reaches the occupied spaces of a building. After exposure to humans and machines, the stale, warm, humid air faces additional resistance as it is pulled back to the AHU through the return air duct system.
HVAC engineers specify the critical parameters of a building's duct system, including the range of airflow volume, temperature, humidity, and pressure that must be present at each point in the system. Air balancers must verify that the HVAC system meets the specifications. They measure the state of the system as installed, issue corrective action requests as necessary, and then tune the system to achieve optimum comfort and energy efficiency within the range of conditions specified.
Air balancers must measure critical air parameters at many key places throughout the building's duct system, including right at the AHU, in the main duct, at the entrance to key branch ducts, such as the ducts feeding each floor of a building, and in the various ducts supplying diffusers in the occupied spaces. The measured air volumes are collected and compared and analyzed in charts to account for every CFM (cubic feet per minute) of air generated by the fan. Leaks are detected and fixed. Rotational speeds (RPM) of fans are adjusted. Valves, dampers, and grilles are adjusted.
Flow and airflow are industry terms that relate to the volumetric rate of fluid flow expressed in units such as cubic feet per minute (CFM). Airflow is usually not measured directly. It is usually calculated by measuring the velocity of air at multiple points in a cross-sectional plane, calculating an average velocity at the plane, and then multiplying by the known area of the cross-section. The plane where measurement takes place might be across an air duct, in a duct-shaped probe like a capture hood, or at the opening of a fume hood, door, or window.
A velocity traverse or “duct traverse” is one of the most complicated and arduous procedure in the air balancing field.
A “duct traverse” is a series of measurements at a particular point in a duct to determine the air volume in cubic feet per minute (“CFM”) and the air velocity profile in feet per minute (“FPM”). It is usually desirable that additional parameters be measured at the same location, including the static pressure, which is the pressure between the air inside the duct and the pressure in the building, and air temperature (dry bulb). Sometimes air moisture content is also measured, in terms of wet bulb temperature or dew point or percent relative humidity or grains of water per cubic foot.
The term “traverse” as used herein means the measurement of every parameter of interest at a particular duct location.
Not only are traverses required in multiple locations of a building, traverses are often required to be performed multiple times over days or weeks at the same location, because the duct system must be tested under various conditions. Nighttime conditions are controlled differently from daytime conditions. Seasons vary and the load of temperature and humidity stress on the building varies. There are often fire and/or smoke control modes with special duct requirements. These requirements mean that duct traverses are among the most frequently performed procedures of air balancing.
Measurement tools and techniques have changed very little in the last century. A Pitot tube is still the velocity probe of choice. The Pitot tube is actually two tubes within a probe shaft that conduct two different air pressures from one end to the other. There are two orifices on one end and two ports on the other end. When the tip of a Pitot tube is properly oriented with its tip facing the direction of airflow, the air colliding with the tip causes Total Pressure, while the air moving parallel to the shaft causes Static Pressure. The mathematical difference between Total Pressure and Static Pressure is called Velocity Pressure. If Velocity Pressure is known, along with temperature and barometric pressure, which determine the density of air, then the following popular equation, derived from fundamental laws of physics, provides the precise velocity of the moving air:V=1096.7×square root of (VP/d), where:
V is velocity in feet per minute
VP is velocity pressure in inches of water column
d is density of air in pounds per cubic foot=1.325×BP/T, where:
BP is barometric pressure in inches of mercury
T is absolute temperature=degrees Fahrenheit+460
A Pitot tube that comprises two tubes, one to conduct total pressure and the other to conduct static pressure is also referred to as a “Pitot-static tube”. The term “Pitot tube” as used herein is intended to be inclusive of the so-called “Pitot-static tube.”
Some velocity probes are based on differential pressure like the 400-year-old Pitot tube, but they have “lee side” orifices instead of static pressure orifices, so the differential pressure generated is not the same as traditional velocity pressure. However, velocity can still be calculated using the equation above, with only the addition of a constant factor K that can be empirically determined such that the equation becomes:V=K×1096.7×square root of (VP/d).
A duct traverse is performed as follows. A technician first measures the length and width of a rectangular duct, or the diameter of a round duct, and calculates the cross-sectional area, adjusting for the thickness of the duct walls and any insulation or other internal obstructions. Then he consults a table provided by an engineering society, such as ASHRAE, for the locations of the points in a matrix on the duct cross-sectional plane at which air velocity must be known in order to make an accurate calculation of average air velocity. The technician drills holes in the duct to allow the Pitot tube to be positioned at the each point in the matrix. It is convenient to think about horizontal and vertical planes across the duct. The technician marks his probe with tape so he can see how far into the duct to insert it to reach each traverse point.
To perform the measurements in a typical non-residential site, the technician usually has to stand on a high ladder, with his head above the ceiling tiles, and balance precariously while manipulating tools in both outstretched hands. In one hand is the meter. Connected to the meter with tubes is a Pitot tube probe, which is inserted into the duct and placed at the point of interest. Holding the Pitot tube as steady as possible, the technician pushes a button on the meter to make a measurement. Then the technician moves the Pitot tube probe to the next point and makes another measurement. The technician must manipulate the meter with one hand to press the control keys while manipulating the Pitot tube with the other hand and keeping the tubes from swinging and getting tangled.
The technician then makes a velocity measurement at each traverse point, one after the other, recording or storing each reading as he goes. Usually, between 16 and 100 measurements are required, depending on the size of the duct, each one taking a few seconds or several seconds. When each point in the matrix has been measured, the Pitot tube is withdraw from the duct. The temperature and/or humidity probe is withdrawn from the duct. The holes are plugged to prevent air leaking out. The average of all measurements is recorded as the average velocity at that duct cross-section. When multiplied by the cross-sectional area, the volumetric airflow is determined.
After the traverse measurements, the technician must measure the static pressure in the duct at that location. Traditional instruments will not allow static pressure to be measured with the same setup as velocity. The technician must remove the Pitot tube from the duct, change the air hose attachments, and change the mode on the meter. The technician then re-inserts the Pitot tube, or a different probe for measuring static pressure, back into the duct. A series of measurements are made to determine the most representative static pressure at that location, and it is noted.
Beyond the basic set-up procedures of determining duct size, determining matrix points, and drilling duct holes, these steps are required: attach temperature probe to meter, set meter to temperature mode, insert temperature probe in duct, measure temperature, change meter mode to velocity, attach pressure tubes to meter and Pitot tube, insert Pitot tube into duct, measure velocity at traverse points and store data in memory for review and statistics, withdraw Pitot tube from duct, change tubes at meter and at Pitot tube for static pressure, change meter mode to static pressure, insert Pitot tube into duct, measure static pressure, remove Pitot tube from duct.
There are well known problems with the duct traverse procedure including, but not limited to the time consuming nature of performing the measurements; the precarious manner in which the measurements are made; and the multiple steps required.
Conventional commercially available instruments are designed to measure and display only one parameter at a time, and require regular manual operation to even do that. A typical hand-held meter or instrument comprises a plastic case enclosing a printed circuit board with microprocessor-controlled electronics, memory, one or more sensors, and a display. A sensing probe is connected using cables, wires, tubes, or other means. Various probes, large and small, are designed to collect environmental samples for sensing, measurement, display, and storage. The user often must wait between 2 and 8 seconds for the meter to generate a reading. The reading is then displayed by the meter and the user can either write it down or store it in the memory of the meter. All of this is required to determine the velocity at a single point in the duct.
Foil types of velocity probes are sometimes preferred because they are easier to insert through a hole into the duct, not having the bend of the traditional Pitot tube. Unlike Pitot tubes, foil-type probes can also measure negative velocity, the velocity of air moving in the opposite direction due to eddies near duct discontinuities. However, the foil-type of probe has a lee-side orifice instead of a true static pressure orifice, so the static pressure measurement of a complete traverse requires the replacement of the velocity probe with a traditional static pressure probe, known as a “static tip”. This requires time and limits the productivity of the technician, who may decide to skip the measurement or estimate the static pressure.
A problem of conventional practice is that the traverse measurements stored in the conventional meter are required at a different physical location. Conventional practice is for the meter to be brought to a table where the stored data is either transferred manually to a computer or report form, or the data is loaded electronically into a personal computer for subsequent report generation.
Typical meters are generally so large and heavy that they require a technician to devote a hand to hold them and another hand to press control keys. With difficulty, a technician learns to hold an instrument in his palm while fingering the keys with the thumb of the same hand. That hand is not available to steady or brace the technician who stands in a precarious situation. Accordingly, there is a need for instruments that are small and light and able to be mounted and supported without requiring a human hand and arm.
Airflow meters provide low velocity accuracy. Typical accuracies are specified as 3%+/−7 fpm. A reading of 500 FPM could really be 478 to 522. But at 100 FPM, the velocity could really be 90 or 110, and at 50 FPM, the velocity could really be 41.5 or 58.5. That not considered accurate enough.
Measurements may be inaccurate for several reasons that are independent of metering accuracy. More specifically, accuracy is lost when physical stress causes the technician to inadvertently move the velocity probe during measurement or hold the probe in the wrong location; accuracy is lost when the long, dangling rubber hoses between the instrument and the probe are allowed to swing, causing waves that affect the pressure sensors; accuracy is lost when the technician rushes through the process, taking too few velocity measurements or taking other shortcuts; and accuracy is lost when duct air temperature is often ignored due to the difficulty and time required to place and hold the temperature probe for a proper measurement. The typical probe is attached to the meter via a coiled cable.
Accuracy is also lost when the velocity profile is not uniform enough to meet industry standards. The industry-prescribed matrix locations were developed over many years and much research to ensure an accurate result of traverses. Industry standards forbid the performance of a traverse in areas where fans, dampers, louvers, duct bends, or other discontinuities, cause air turbulence and uneven airflow. In the proscribed sections of duct, air eddies and reverse currents can exist. Velocity traverses in these areas, if conducted on the standard matrix, will not be accurate. It is necessary that air passing through such discontinuities be allowed to even out over many feet of straight duct, after which a typical velocity profile is achieved. However, the reality is that architects and engineers are not required to provide such a proper location for a duct traverse, and they often cannot be found. In these cases technicians are forced to measure at the undesirable profile point. The accuracy of the average velocity calculated would be improved if the number of measurement points in the matrix were increased substantially. Current industry standards allow that—specifications are for the minimum number of readings. However, technicians are reluctant to do that because the procedure is already so time-consuming.